Detection method for change of high-consequence areas based on multi-source remote sensing image

With the continuous promotion of urbanization in China in recent years, the surrounding environment of some pipelines is changing constantly, and the scope of the high-consequence areas of the pipeline section is also changing. Accurately and efficiently mastering the data of environment change to update the information of the high-consequence areas is very important for the safety management of the pipeline companies. According to the characteristics of less change of natural features but frequent change of artificial features along the pipeline in short term, a method for detection of the change of the high-consequence areas based on the multi-source remote sensing image with high-resolution orthoimage and multispectral image was put forward. In this method, the change information of multi-source data is extracted with the LBP-HOG fusion feature and spectral gradient difference method based on the super-pixel segmentation. Experiments show that, the change of artificial features along the pipeline can be detected accurately with this method, so that the efficiency for management of high-consequence areas could be improved effectively.